[Not Applicable]
[Not Applicable]
Certain embodiments of the present invention relate to a medical diagnostic ultrasound system for the purpose of, for example, imaging tissue structure within a human subject. More particularly, certain embodiments relate to a method and apparatus for generating compounded image frames of tissue structure from multiple ultrasound beams along intersecting scan lines in an image plane.
Ultrasound systems have been proposed that employ compounding techniques that combine information sampled from different data frames. For example, the combining technique may include summing or averaging demodulated data from multiple data frames. Several consecutive frames of data are sampled with each frame of data typically being gathered at a unique steering angle. The steering angle is the angle that an ultrasound beam makes in a scan plane with respect to the surface of the ultrasound transducer. The frames of data are then compounded (combined) to create an image frame for display.
Because the data frames are each gathered at a unique steering angle, more information is effectively gathered per sample volume location within a scan plane since reflections from a given sample volume are being viewed at different aspects due to the multiple steering angles. This is analogous to a person viewing a building from significantly different aspects or angles. At each aspect, the person observes different features of the building. There may even be an aspect where the building is blocked and cannot be seen at all. Hopefully, other aspects provide legitimate viewing of the building.
Similarly, for a sample volume of tissue within a scan plane, ultrasound reflections from each aspect will be different, providing different amplitude and phase information back to the ultrasound system. As a result, an image frame with more tissue detail than a frame formed at a single steering angle is provided.
Unfortunately, the resultant compounded image frame may be corrupted due to motion of the tissue during the time delay between successive frames of sampled data. Each frame of data is composed of data gathered from sample volumes of tissue within a scan plane. When motion occurs between one data frame and the next, due to the subject moving with respect to the ultrasound transducer, then data gathered from a given sample volume may appear in a different location in one frame compared to another frame. During compounding of the data frames, data from locations representing different tissue sample volumes will be combined with each other due to the motion. Motion artifacts may result in the compounded image that appear as blurred sample volumes, thus causing detailed tissue information to be lost. The sample volumes in the image lose their sharpness. Hence, the detail of tissue, which could be seen in a compounded image without motion artifact, is lost. Frame-to-frame motion compensation techniques have been employed before compounding to try to reduce motion artifacts. However, the frame-to-frame motion compensation techniques proposed thus far are time consuming, thus degrading frame rate, and sometimes do not compensate well for frame-to-frame motion.
Conventional compounding and scanning techniques are disclosed in the following patents. U.S. Pat. No. 4,649,927 to Fehr et al. is directed to methods and apparatus for producing a compound ultrasound image from overlapping body scans. U.S. Pat. No. 4,319,489 to Yamaguchi et al. is directed to combining images obtained through normal linear scanning and inclined linear scanning. U.S. Pat. No. 6,117,081 to Jago et al. is directed to correcting mis-registered images that are to be spatially compounded. U.S. Pat. No. 4,159,462 to Rocha et al. is directed to performing overlapping sector scans.
A need exists for an approach to generating a compounded image frame having enhanced image quality and reduced motion artifacts without having to perform motion compensation. A need also exists to achieve reduced-artifact compounded imaging at frame rates that are acceptable for abdominal scanning in real-time.
An embodiment of the present invention provides an ultrasound system for imaging structure within a subject by generating a compounded image frame having reduced motion artifacts. The system transmits ultrasound waves into the structure and, for at least one sample volume location within an image plane of the structure, forms a set of received beams in response to ultrasound waves backscattered from the structure such that the set of received beams intersect at the sample volume location. A compounded data value is generated corresponding to the at least one sample volume location within the image plane based on the corresponding set of received beams. A compounded image frame is formed from at least one compounded data value.
Apparatus is provided for compounding data values sampled by the ultrasound machine at sample volume locations. The term xe2x80x9ccompoundingxe2x80x9d as used throughout means combining multiple data values, either coherently or non-coherently, to create a new, single data value. The apparatus includes a beamformer and transducer array for generation of multiple beams at different steering angles and originating from the same point near the surface of the transducer array. A single data frame is formed by duplicating the formation of the beams such that the beams originate at different points across the lateral dimension of the transducer array. The apparatus further includes a data compounding module to adaptively weight and filter data values from intersecting beams to create a compounded data value corresponding to a sample volume in the scanned image plane. In other words, the compounded image is generated using only the data gathered in a single data frame. One compounded image frame is formed from one data frame. Motion compensation is not required because the intersecting beams for a given sample volume are sampled close enough in time to avoid significant motion artifacts. The apparatus generate a plurality of compounded image frames at a frame rate of at least 4 frames per second for real-time imaging of the structure within the subject.
A method is also provided for compounding data values sampled by an ultrasound machine at sample volume locations within an image plane. The method includes generating a set of multiple beams at different steering angles that originate from the same point at the top of the image plane. A single data frame is formed by duplicating the method at different points across the lateral dimension of the image plane. The method further includes data compounding to adaptively weight and filter data values from intersecting beams to create a compounded data value corresponding to a sample volume in the scanned image plane. In other words, the compounded image is generated using only the data gathered in a single data frame. One compounded image frame is formed from one data frame. Motion compensation is not required because the intersecting beams for a given sample volume are sampled close enough in time to avoid significant motion artifact.
Certain embodiments of the present invention afford an approach to generating a compounded image frame having reduced motion artifact without having to perform motion compensation. Reduced-artifact compounded imaging at frame rates that are acceptable for abdominal scanning in real-time is also achieved.